List your company
ResearchFormerly WizWand

Algorithmic Reasoning on MATH (Accuracy)

80.6AccuracyRM-Regen
25.68839.94454.268.456Mar 20, 2026
Updated 7mo ago

Evaluation results

MethodLinks
RM-RegenModel=Llama3.1-8B, Verification Setting=Oracle2026.0380.6
Reflexion(3 iters)Model=Llama3.1-8B, Verification Setting=Oracle2026.0377.8
Best-of-N (N=3)Model=Llama3.1-8B, Verification Setting=Oracle2026.0376.8
ReflectEvoModel=Llama3.1-8B, Verification Setting=Oracle2026.0373.6
ST CoTBase Model=Llama 3.1-8B, Iterations=22026.0372.2
RM-RegenModel=Gemma2-9B, Verification Setting=Oracle2026.0372.2
Reflexion(3 iters)Model=Gemma2-9B, Verification Setting=Oracle2026.0372
RM-RegenBase Model=Llama 3.1-8B2026.0371.4
RM-RegenModel=Llama3.1-8B2026.0371.4
ST CoTBase Model=Llama 3.1-8B, Iterations=32026.0369.8
ST CoTModel=Llama3.1-8B, iterations=32026.0369.8
ProCoBase Model=Llama 3.1-8B, Iterations=22026.0369.6
ProCoBase Model=Llama 3.1-8B, Iterations=32026.0369.6
ProCoBase Model=Llama 3.1-8B, Iterations=52026.0369.6
ProCoModel=Llama3.1-8B, iterations=32026.0369.6
ProCoBase Model=Llama 3.1-8B, Iterations=42026.0368.2
Best-of-N (N=3)Model=Gemma2-9B, Verification Setting=Oracle2026.0367.2
ST CoTBase Model=Llama 3.1-8B, Iterations=42026.0366.4
RM-RegenModel=Gemma2-9B2026.0366.2
ProCoModel=Gemma2-9B, iterations=32026.0364.6
Best-of-N (N=3)Model=GPT-3.5, Verification Setting=Oracle2026.0364
ReflectEvoModel=Gemma2-9B, Verification Setting=Oracle2026.0363.8
RM-RegenModel=GPT-3.5, Verification Setting=Oracle2026.0363.4
ST CoTBase Model=Llama 3.1-8B, Iterations=52026.0363
Reflexion(3 iters)Model=GPT-3.5, Verification Setting=Oracle2026.0362.25
ST CoTModel=Gemma2-9B, iterations=32026.0359.4
RM-RegenBase Model=GPT-3.52026.0356.8
RM-RegenModel=GPT-3.52026.0356.8
ST CoTBase Model=GPT-3.5, Iterations=22026.0354.8
Self-RefineBase Model=Llama 3.1-8B, Iterations=22026.0354.8
ProCoBase Model=GPT-3.5, Iterations=22026.0354.6
ST CoTBase Model=GPT-3.5, Iterations=32026.0353.6
ST CoTModel=GPT-3.5, iterations=32026.0353.6
Self-RefineBase Model=Llama 3.1-8B, Iterations=42026.0353.2
ST CoTBase Model=GPT-3.5, Iterations=42026.0353
ProCoBase Model=GPT-3.5, Iterations=32026.0352.2
ProCoModel=GPT-3.5, iterations=32026.0352.2
ProCoBase Model=GPT-3.5, Iterations=42026.0352.08
Self-RefineBase Model=Llama 3.1-8B, Iterations=32026.0351
Self-RefineModel=Llama3.1-8B, iterations=32026.0351
Self-RefineBase Model=Llama 3.1-8B, Iterations=52026.0349.6
Self-RefineBase Model=GPT-3.5, Iterations=42026.0348.8
Self-RefineBase Model=GPT-3.5, Iterations=22026.0348.2
Self-RefineBase Model=GPT-3.5, Iterations=32026.0347.2
Self-RefineModel=GPT-3.5, iterations=32026.0347.2
Self-RefineModel=Gemma2-9B, iterations=32026.0327.8